Effect of extrinsic factors on the detection of PRRSV and a porcine-specific internal sample control in serum, oral fluid, and fecal specimens tested by RT-rtPCR

Abstract

We characterized the effect of 1) temperature × time, 2) freeze-thaw cycles, and 3) high porcine reproductive and respiratory syndrome virus (PRRSV) RNA concentrations on the detection of PRRSV and a porcine-specific internal sample control (ISC) in serum, oral fluid, and fecal samples using a commercial PRRSV RT-rtPCR assay (Idexx). In study 1, the effect of temperature × time on PRRSV and ISC detection was shown to be specimen dependent. In serum stored at 4, 10, or 20°C, PRRSV detection was consistent for up to 168 h, but storage at 30°C reduced detectable PRRSV RNA. ISC RNA was stable in serum held at 4 and 10°C, but not at 20 and 30°C. In contrast, PRRSV and ISC RNAs in oral fluid and fecal samples continuously decreased at all temperature × time treatments. Based on these data, serum samples should be stored at ≤ 20°C to optimize PRRSV RNA detection. Oral fluid and fecal samples should be frozen in a non–self-defrosting freezer until tested. In study 2, freeze-thaw cycles had little impact on PRRSV and ISC detection, but more so in oral fluids than serum or fecal samples. Thus, freeze-thaw cycles in oral fluids should be minimized before RT-rtPCR testing. In study 3, the ISC was not affected by high concentrations of PRRSV RNA in serum, oral fluid, or fecal samples. It should not be assumed that data from our PRRSV study are applicable to other pathogens; additional pathogen-specific studies are required.

In molecular testing, commercial assays include quality controls to monitor for technical errors in the testing process (e.g., positive and negative extraction controls, positive and negative amplification controls). In addition, PCR assays may also monitor reference genes inherent to the specimen (i.e., internal sample controls, ISCs) to verify that nucleic acid (NA) integrity was maintained throughout the process of sample collection, transport, and real-time PCR (rtPCR) testing.^8,21 In human laboratory medicine, examples of genes used as ISCs include glyceraldehyde-3-phosphate dehydrogenase (GAPDH) for the detection of hepatitis C virus (HCV; Hepacivirus hominis) ³⁰ and β-glucuronidase for the detection of human influenza A and B viruses (Alphainfluenzavirus influenzae, Betainfluenzavirus influenzae, respectively). ¹² Research in veterinary laboratory medicine has reported the use of species-specific ISCs in an avian influenza A virus reverse-transcription rtPCR (RT-rtPCR) assay (β-actin gene), ³⁶ African swine fever virus rtPCR assay (β-actin gene), ³⁷ and a PCR assay for the detection of Salmonella enterica in cattle. ³

Initially identified in 1991, porcine reproductive and respiratory syndrome virus (PRRSV; Betaarterivirus suid 1 and 2) continues to impose major costs on the global swine industry.^17,25,35 Strategies for PRRSV prevention and control invariably include diagnostic and surveillance testing to establish the infection status of individual pigs and/or groups of animals. In 2021, specimens submitted for PRRSV RT-rtPCR testing at 5 midwestern U.S. veterinary diagnostic laboratories included oral fluid (39%), processing fluid (19%), serum (16%), miscellaneous specimen types (12%), lung (8%), and blood swabs (6%; Trevisan G, pers. comm., 2022 Oct 25).

Recommendations for sample handling to optimize PRRSV detection include promptly chilling or freezing after collection on the farm and maintaining the cold chain through arrival at the diagnostic laboratory. ² However, it is not always logistically possible to meet these recommendations in the field, and samples may have been subjected to inappropriate temperatures or undergone multiple freeze-thaw cycles between the time of sample collection and arrival at the laboratory. Broadly speaking, RNA is more susceptible to degradation than DNA given the presence of a hydroxyl group that makes it prone to hydrolysis. ²⁶ For example, a study demonstrated that total NA recovery from reference genes in mummified tissue was more challenging in β-actin RNA (8 of 12 specimens) than in GAPDH DNA (12 of 12 specimens). ²²

Undegraded NA is mandatory for the optimal performance of PCR-based testing. ⁵ Although there are several reports on the stability and/or infectivity of swine viruses,^{10,11,18,19,23,29} there is limited information on the effect of these conditions on RT-rtPCR performance. Additionally, although ISCs potentially offer the opportunity to document the suitability of samples for rtPCR testing, we found no refereed publications on the robustness or constancy of ISCs in specimen matrices.

To address this information gap, we characterized the effect of extrinsic factors (i.e., temperature × time, freeze-thaw) on the detection of PRRSV and a porcine-specific ISC in serum, oral fluid, and fecal samples using a commercial PRRSV RT-rtPCR assay that detects both targets simultaneously (RealPCR NA PRRSV type1 and type 2 multiplex RNA mix; Idexx). The kit insert provided with the assay identifies the gene targeted by the ISC as host-endogenous RNA but does not specify the targeted gene (proprietary). In addition, we evaluated the effect of high concentrations of PRRSV RNA on the detection of the porcine-specific ISC in the 3 specimen types to test for possible competitive interference.

Materials and methods

Experimental design

We conducted 3 studies (Table 1) to characterize the detection of PRRSV and ISC RNAs using a commercial PRRSV RT-rtPCR assay (Idexx). In study 1, we quantified the effect of temperature and time on RT-rtPCR results; in study 2, we tested the effect of repeated freeze-thaw cycles; in study 3, we evaluated the impact of high concentrations of PRRSV on the detection of the ISC RNA. For data analyses, PRRSV and ISC RT-rtPCR quantification cycles (Cqs) were re-expressed as “efficiency standardized” Cqs (ECqs).^7,27 Thereafter, ECqs were transformed (cube root), and the data was analyzed using a mixed-effects regression model (MRM; R v.4.2.1, https://www.r-project.org/).

Table 1.

Overview of 3 studies on the effect of extrinsic factors on the detection of porcine reproductive and respiratory syndrome virus (PRRSV) and a porcine-specific internal sample control (ISC) in serum, oral fluid, and fecal samples using a PRRSV RT-rtPCR assay (Idexx).

Study	Source of specimens	Experimental design
Study 1. Temperature by time effect	Sera from 5 pigs IN inoculated with wild-type PRRSV2 isolates (28 dpi). Oral fluid and feces from 5 vaccinated pigs (Ingelvac PRRS MLV; 5, 6, 8, 9, 14 dpv).	Aliquots of 28 individual pig specimens assigned to 1 of 28 combinations of storage temperature (4°, 10°, 20°, or 30°C) × time (24, 48, 72, 96, 120, 144, or 168 h).
Study 2. Freeze-thaw effect	Positive sera from 5 pigs IN inoculated with 5 different PRRSV2 field isolates (28 dpi). Negative serum from 5 mock-inoculated pigs. Oral fluid and feces from 10 pigs before and after vaccination (Ingelvac PRRS MLV; −7, 5, 6, 8, 9, 14, 35 dpv).	Aliquots of 4 individual pig specimens exposed to 1 of 4 freeze-thaw treatments (2, 5, 10, or 15 cycles).
Study 3. Effect of high concentration of PRRSV RNA on the detection of ISC RNA	PRRSV-negative serum (5 mL), oral fluid (5 mL), and fecal suspension (5 mL) from pigs housed in the Iowa State University research facilities from experiments not involving PRRSV.	PRRSV MLV (Ingelvac PRRS MLV vaccine) resuspended with 5 mL of a PRRSV-negative matrix (serum, oral fluid, fecal suspension) and tested at six 10-fold serial dilutions.

dpi = days post-inoculation; dpv = days post-vaccination; IN = intranasally.

Specimens for study 1 and study 2

Serum samples were from a study comprising 72 PRRSV-negative 3-wk-old pigs randomly allocated to 1 of 6 groups of 12 pigs each using a random number generator (Excel; Microsoft). As described elsewhere, ³¹ each group was housed in a separate room under biosafety level 2 (BSL-2) conditions at animal research facilities at Iowa State University (ISU; Ames, IA, USA). After 1 wk of acclimation, 8 of these pigs were inoculated intramuscularly (2 mL/pig) and intranasally (2 mL/naris) with 1 of 4 PRRSV 1-4-4 isolates (L1C-variant, L1C non-variant, L1A, or L1H) or a PRRSV 1-7-4 isolate at a concentration of ~1 × 10⁶ TCID₅₀ per mL. Pigs in the negative control group were mock inoculated with medium (Roswell Park Memorial Institute [RPMI] 1640 medium; Gibco, Thermo Fisher) by the same routes and volumes. Contact pigs (n = 4) were introduced into each group at 2 d post-inoculation (dpi). Serum samples used in studies 1 and 2 were collected at 28 dpi (i.e., 1 pig in each group from the 5 PRRSV-inoculated groups, and 5 pigs from the negative control group). In brief, pigs were bled from the jugular vein using a single-use vacutainer (10-mL SST Vacutainer; Becton Dickinson). Serum samples were obtained by centrifuging blood at 1,000 × g for 10 min and were stored at −80°C until used. Before initiating the experiment, samples were confirmed positive for PRRSV by RT-rtPCR assay (Idexx). At the end of the experiment, pigs were euthanized by chemical injection, and large volumes of blood were collected by exsanguination.

Oral fluid and fecal samples were from a study involving 12 PRRSV-negative 14-wk-old pigs housed individually in pens within a BSL-2 research facility at ISU, described elsewhere. ¹⁶ After 5-d of acclimation, pigs were vaccinated intramuscularly with a PRRSV modified-live virus (MLV) vaccine (2 mL, Ingelvac PRRS MLV; Boehringer Ingelheim). Oral fluid samples were collected from individual pigs daily from −7 until 42 d post-vaccination (dpv) using cotton rope (Web Rigging Supply). Immediately thereafter, samples were aliquoted into 5-mL tubes (Fisher Scientific) and stored at −80°C. Similarly, individual pig fecal samples were collected from the floor of each pen each morning from −7 to 42 dpv. In the laboratory, samples were aliquoted into 50-mL tubes (Corning; Thermo Fisher) and stored at −80°C. We used a subset of the samples collected in the study, including PRRSV-negative oral fluids collected before vaccination, PRRSV-negative feces from 35 dpv, and PRRSV-positive oral fluids and feces from 5, 6, 8, 9, and 14 dpv.

Specimens for study 3

All serum, oral fluid, and fecal specimens used in study 3 were PRRSV-negative. The serum originated from an experiment involving porcine epidemic diarrhea virus in multiparous gestating sows followed from −1 to 12 dpi, as described elsewhere. ²⁸ Animals were housed at the ISU Livestock Infectious Disease Isolation Facility (Ames, IA, USA). At the termination of the project, pigs were euthanized by chemical injection and exsanguinated. Blood samples were processed by centrifugation at 1,500 × g for 15 min and stored at −20°C until processed.

Oral fluids and fecal specimens were from pigs in the negative control group in a study not involving PRRSV. ⁶ Animals were housed in individual pens at the ISU Livestock Infectious Disease Isolation Facility. Oral fluid samples were collected from individual pigs twice daily (morning and afternoon) for 49 d using cotton rope (Web Rigging Supply) and then aliquoted into 50-mL tubes (Falcon; Fisher Scientific) for storage at −20°C. Fecal samples were collected from the floor of each pen twice daily and then aliquoted into 50-mL tubes for storage at −20°C.

PRRSV RT-rtPCR reference standards

Specimen- and target-specific reference standards were created by rehydrating a 10-dose vial of lyophilized PRRSV MLV vaccine (Ingelvac) with 20 mL of the respective PRRSV-negative specimen (serum, oral fluid, or fecal suspension). The vaccine suspension was then 10-fold diluted and the 1 × 10⁴ dilution used as the reference standard. Four reference standard samples were included in each RT-rtPCR run and their testing results were used to convert PRRSV and ISC Cqs to ECqs, as described in the PRRSV RT-rtPCR experimental design section.

Study 1

To test the effect of temperature by time, 5 PRRSV-positive serum samples, 5 oral fluid samples, and 5 fecal samples (Table 1) were divided into 28 aliquots of 500 µL and placed in uniquely numbered 2-mL tubes (Cryogenic vials; Corning). Aliquots of fecal samples were created by placing 1 g of feces and 1 mL of PBS (Gibco; Thermo Fisher), i.e., 50% w/v suspension, in the 2-mL tubes, and vortexing for 20 s.

Each numbered aliquot was randomly assigned (https://random.org) to one of 28 combinations of temperature × time treatment (i.e., 4, 10, 20, or 30°C for 24, 48, 72, 96, 120, 144, or 168 h). Treatments at 4, 10, and 20°C were conducted in temperature-controlled environmental chambers (6010-1 environmental test chamber; Caron Products). Treatment at 30°C was done in a water-jacketed CO₂ incubator (NAPCO 6301; Thermo Fisher). Temperature was monitored continuously throughout the treatment period (Fisherbrand traceable digital thermometer; Fisher Scientific) and shown to be stable; temperatures were 4°C (3.9–4.1°C), 10°C (10–10.1°C), 20°C (19.9−20.4°C), and 30°C (29.8–30°C) over the course of the study. As the treatment time was completed, 140 samples per specimen were removed and stored at −80°C until all treatments were completed. Samples were then randomized within specimen and tested by RT-rtPCR assay for PRRSV and ISC RNAs.

Study 2

To evaluate the effect of freeze-thaw cycles, serum samples (5 PRRSV-positive, 5 PRRSV-negative), oral fluids (6 PRRSV-positive, 4 PRRSV-negative), and fecal samples (5 PRRSV-positive, 5 PRRSV-negative; Table 1) were divided into 4 aliquots of 1 mL each and placed in 2-mL tubes (Cryogenic vials; Corning). As in study 1, PBS was added to feces to create a 50% w/v suspension.

Aliquots were exposed to 1 of 4 freeze-thaw treatments (2, 5, 10, or 15 cycles). Freeze-thaw cycles consisted of placing samples at −80°C each morning and then placing at 4°C overnight. Aliquots (40 per specimen) were stored at −80°C 1 cycle before completing their respective freeze-thaw treatment (e.g., samples to undergo 5 freeze-thaw treatments were stored at −80°C after 4 freeze-thaws). Thereafter, samples were randomly ordered within specimen, thawed at 4°C, and tested in triplicate (120 per specimen) with the PRRSV RT-rtPCR assay (Idexx) for PRRSV and ISC RNA detection.

Study 3

To evaluate the effect of high PRRSV RNA concentrations on ISC detection, six 10-fold dilutions of a PRRSV MLV vaccine (Ingelvac) were created using PRRSV-negative serum, oral fluid, and feces (20% w/v fecal suspension) as diluent. The fecal suspension was created by placing 10 g of feces and 50 mL of PBS in a 50-mL tube (Nunc; Thermo Fisher). Thereafter, the sample was vortexed vigorously for 20 s and centrifuged (3,300 × g for 3.5 h). The supernatant (fecal suspension) was aliquoted into a clean 50-mL tube.

To create high PRRSV RNA concentrations, rather than resuspending a 10-dose vial of PRRSV MLV vaccine (Ingelvac) with 20 mL of sterile diluent, the lyophilized vaccine virus was resuspended with 5 mL of serum, oral fluid, or the 20% fecal suspension. Thereafter, the vial was incubated at 25°C for 10 min to completely dissolve the lyophilizate. Six 10-fold dilutions (10^-1–10^-6) were then created using the PRRSV-negative specimen (serum, oral fluid, or fecal suspension) as diluent. All 10-fold dilutions were randomized within specimen and tested in triplicate using the PRRSV RT-rtPCR assay (Idexx).

RNA extraction

All laboratory procedures were performed in a certified biological safety cabinet using in-date calibrated pipettes. Extraction of RNA from serum, oral fluid, and fecal samples was done (RealPCR DNA/RNA spin column kit; Idexx) following the manufacturer’s instructions. Before extraction, fecal samples were vortexed for 20 s, centrifuged at 8,000 × g for 30 s, and RNA was extracted from the supernatant.

For serum, the lysis working solution was prepared with 195 µL of lysis buffer, and 5 µL of carrier RNA per sample. Thereafter, 5 µL of proteinase K was mixed with 200 µL of the sample in a 2-mL tube (Idexx). The mixture was vortexed for 10 s, centrifuged at 8,000 × g for 30 s, and 200 µL of lysis working solution was added. For oral fluid and fecal samples, 190 µL of lysis buffer, 5 µL of carrier RNA, and 5 µL of proteinase K were mixed in a 15-mL tube (Thermo Fisher) and used as lysis working solution. Thereafter, 200 µL of the lysis working solution was added to a 2-mL tube followed by addition of 200 µL of each sample.

All subsequent procedures were identical for serum, oral fluids, and fecal samples. Samples mixed with their respective lysis working solution were incubated for 3 min at 25°C and centrifuged at 8,000 × g for 30 s. Subsequently, 200 µL of ethanol was added and then the mix was vortexed for 10 s, incubated for 5 min at 25°C, centrifuged at 8,000 × g for 30 s, and 600 µL were transferred to a spin column with a silica membrane to bind genetic material. Thereafter, columns were centrifuged at 8,000 × g for 3 min; washed with 400 µL of wash solution 1, 400 µL of wash solution 2; and centrifuged between each wash at 11,000 × g for 30 s. Columns were washed a third time with 200 µL of wash solution 2 and centrifuged at 20,000 × g for 2 min. Then, NAs were eluted from the columns with 50 µL of 70°C elution water, incubated for 1 min at 25°C, centrifuged at 20,000 × g for 1 min, and collected into 2-mL tubes for immediate RT-rtPCR testing. Positive and negative extraction control samples were included in every extraction run.

PRRSV RT-rtPCR assay

The PRRSV RT-rtPCR assay (Idexx) was performed as directed by the manufacturer. The extracted NAs (5 µL) were added to the master mix (10 µL) and the RNA target mix (10 µL) containing primers and probes for both PRRSV and the ISC. Thereafter, the RT-rtPCR run was performed in a magnetic induction cycler (Mic qPCR cycler; Bio Molecular Systems) as directed by the manufacturer: 45 cycles of reverse transcription (RT) at 50°C for 15 min, denaturation at 95°C for 15 s, amplification at 95°C for 15 s, and a second amplification step at 60°C for 30 s. Quality controls included in each run (≤ 48 tubes) consisted of positive (RealPCR positive control; Idexx) and negative (RealPCR PCR grade water; Idexx) amplification controls plus 4 specimen-specific, target-specific reference standards.

Results were read with the Mic qPCR cycler software (v.2.10.4; Bio Molecular Systems) and reported as Cqs. Thereafter, PRRSV and ISC RT-rtPCR Cqs were re-expressed as ECqs, as given in equation 1.^7,27

$$\mathrm{E}\mathrm{C}\mathrm{q}={\mathrm{E}}^{-\mathrm{\Delta }\mathrm{C}\mathrm{q}}={\mathrm{E}}^{-(\mathrm{C}\mathrm{q}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e}-\mathrm{m}\mathrm{e}\mathrm{a}\mathrm{n}\mathrm{C}\mathrm{q}\mathrm{o}\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{f}\mathrm{e}\mathrm{r}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{s})}$$ (1)

Amplification efficiency can be expressed either as a ratio (number of target amplicons at the end of a PCR cycle divided by the number at the beginning) or as a percentage. Thus, a doubling at each cycle would represent an efficiency of 2 or 100%. In equation 1, E is the mean amplification efficiency of the 4 reference standards expressed as a ratio, and ΔCq is the difference between a sample’s Cq and the mean Cq of the 4 reference standards included in the run. In our study, amplification efficiency (%) was calculated by the Mic qPCR cycler software (v.2.10.4) using equation 2, with values between 0 and 1 (i.e., 0–100%).

$$\mathrm{E}={10}^{-1/\mathrm{s}\mathrm{l}\mathrm{o}\mathrm{p}\mathrm{e}}-1$$ (2)

For use in equation 1, amplification efficiency (%) was then re-expressed as the fold increase per cycle by adding 1 to the mean amplification efficiency value of the 4 reference standards.

Data analysis

Raw Cqs were re-expressed as ECqs (equation 1) to account for variation in amplification efficiency and to normalize sample results in the context of run-specific reference standards.^7,32,38 Essentially, ECqs express the concentration of target in the sample relative to the concentration of target in the reference standards included in the RT-rtPCR run (i.e., relative quantification). For data analysis, ECq values were transformed to their cube root and then analyzed for studies 1–3 using the “lme4” package in R v.4.2.1. For study 1, the effect of storage temperature × time on PRRSV and ISC RNAs was quantified using a mixed effects regression model. An initial model with “specimen”, “temperature”, “time”, and “temperature*time” as fixed effects, and “sample ID” as the random effect was fitted to evaluate the overall significance of each fixed effect on PRRSV and ISC ECqs using a type-III ANOVA. Thereafter, an MRM was fitted for each specimen (serum, oral fluid, fecal samples) by temperature (4, 10, 20, 30°C) with “time” as fixed effect and “sample ID” as the random effect. The slope of the regression line from each model was used to calculate the estimated PRRSV and ISC ΔECq every 24 h, per temperature, per specimen.

For study 2, an MRM was fitted with “freeze-thaw treatment” and “specimen” as fixed-effects, and “sample ID” as random effect. A type-III ANOVA was used to assess the overall difference of freeze-thaw treatments on PRRSV and ISC RNAs between specimens. The slope of the regression line was used to estimate the expected loss in PRRSV and ISC ECqs per freeze-thaw cycle on each specimen.

In study 3, the effect of PRRSV concentration on the ISC RNA was evaluated for each specimen (serum, oral fluid, fecal suspension) using an MRM with PRRSV concentration (PRRSV ECq) and “specimen” as fixed effects, and “sample ID” as random effect. The effect was also compared between specimens using a type-III ANOVA of the fitted regression model. The slope of the regression line was used to calculate the estimated change (Δ) in ISC ECqs as a function of PRRSV concentration (PRRSV ECqs) per specimen.

Results

Temperature × time treatment (study 1) affected both PRRSV and ISC ECqs in serum, oral fluid, and fecal samples (type-III ANOVA: p < 0.05), with significant differences between specimens (type-III ANOVA: p < 0.05). In serum, PRRSV ECqs were stable at 4, 10, and 20°C (MRM: p > 0.05), but a temperature × time effect was detected at 30°C (MRM: p < 0.05; Table 2, Fig. 1). ISC ECqs in serum had a significant temperature × time effect at 20 and 30°C (MRM: p < 0.05; Table 2, Fig. 2). In oral fluids, PRRSV (Fig. 1), and ISC (Fig. 2) ECqs declined in all temperature × time treatments (MRM: p < 0.05; Table 2). Similarly, in fecal samples, all treatments affected PRRSV (Fig. 1) and ISC (Fig. 2) ECqs (MRM: p < 0.05; Table 2).

Table 2.

Effect of temperature × time (study 1) on porcine reproductive and respiratory syndrome virus (PRRSV) RNA and a porcine-specific internal sample control (ISC) RNA in serum, oral fluid, and fecal samples held at 4, 10, 20, or 30°C for up to 168 h.

Target	Specimen	Temperature	Intercept*	Slope† (95% CI)	p	ΔECq/24 h
PRRSV	Serum	4°C	1.89	0.001 (−0.000, 0.003)	>0.05	0.03
		10°C	2.03	0.000 (−0.002, 0.003)	>0.05	0.00
		20°C	1.97	−0.001 (−0.003, 0.001)	>0.05	−0.02
		30°C	2.00	−0.004 (−0.007, –0.002)	<0.05‡	−0.11
	Oral fluid	4°C	0.53	−0.001 (−0.002, –0.001)	<0.05‡	−0.03
		10°C	0.56	−0.002 (−0.003, –0.001)	<0.05‡	−0.05
		20°C	0.53	−0.003 (−0.004, –0.002)	<0.05‡	−0.07
		30°C	0.46	−0.003 (−0.004, –0.002)	<0.05‡	−0.08
	Feces	4°C	0.84	−0.002 (−0.003, –0.001)	<0.05‡	−0.05
		10°C	0.76	−0.002 (−0.004, –0.001)	<0.05‡	−0.06
		20°C	0.62	−0.004 (−0.005, –0.002)	<0.05‡	−0.09
		30°C	0.58	−0.004 (−0.005, 0.002)	<0.05‡	−0.09
ISC	Serum	4°C	0.76	0.000 (−0.001, 0.002)	>0.05	0.01
		10°C	1.84	−0.000 (−0.002, 0.001)	>0.05	−0.01
		20°C	1.87	−0.003 (−0.004, –0.001)	<0.05‡	−0.07
		30°C	1.77	−0.004 (0.006, –0.002)	<0.05‡	−0.10
	Oral fluid	4°C	0.46	−0.001 (−0.001, –0.000)	<0.05‡	−0.02
		10°C	0.46	−0.002 (0.002, –0.001)	<0.05‡	−0.04
		20°C	0.42	−0.002 (−0.003, –0.002)	<0.05‡	−0.05
		30°C	0.37	−0.002 (−0.003, –0.002)	<0.05‡	−0.05
	Feces	4°C	1.53	−0.002 (−0.003, –0.001)	<0.05‡	−0.06
		10°C	1.44	−0.002 (−0.004, –0.001)	<0.05‡	−0.06
		20°C	1.09	−0.003 (−0.005, –0.001)	<0.05‡	−0.08
		30°C	1.23	−0.004 (0.006, –0.002)	<0.05‡	−0.10

PRRSV and ISC RT-rtPCR (Idexx) results were re-expressed as efficiency standardized Cqs (ECqs) and analyzed by a mixed-effects regression model.

^* Mean ECqs in untreated samples.

^† Mean ECq change per unit increase in time (h).

^‡ Significant temperature × time effect at α = 0.05.

Figure 1.

Regression analyses (mixed-effects regression model, MRM) showing the effect of temperature × time on porcine reproductive and respiratory syndrome virus (PRRSV) RNA in (A) serum, (B) oral fluid, and (C) fecal samples stored at 4, 10, 20, or 30°C for up to 168 h (study 1) and then tested with a PRRSV RT-rtPCR (Idexx). RT-rtPCR results were re-expressed as efficiency standardized Cqs (ECqs). Equations marked with an asterisk (*) indicate that temperature × time had a significant effect on PRRSV ECqs (MRM: p < 0.05).

Figure 2.

Regression analyses (mixed-effects regression model, MRM) showing the effect of temperature × time on a porcine-specific internal sample control (ISC) RNA in (A) serum, (B) oral fluid, and (C) fecal samples stored at 4, 10, 20, or 30°C for up to 168 h (study 1) and then tested with a PRRSV RT-rtPCR (Idexx). RT-rtPCR results were re-expressed as efficiency standardized Cqs (ECqs). Regression equations marked with an asterisk (*) indicate that temperature × time had a significant effect on ISC ECqs (MRM: p < 0.05).

There was an overall difference in the freeze-thaw effect in study 2 for both PRRSV and ISC RNAs between specimens (type-III ANOVA: p < 0.05). In serum, the effect of freeze-thaw treatments was negligible for both PRRSV and ISC RNA (MRM: p > 0.05). In oral fluids, PRRSV and ISC RNAs were affected by freeze-thaw treatments (MRM: p < 0.05); the slope of the regression line had a decay of 0.03 and 0.05 ECqs per freeze-thaw cycle for PRRSV and ISC, respectively. In fecal samples, no freeze-thaw effect was observed in PRRSV or the ISC RNAs for any of the treatments (MRM: p > 0.05; Table 3, Fig. 3).

Table 3.

Effect of freeze-thaw (study 2) on porcine reproductive and respiratory syndrome virus (PRRSV) RNA and a porcine-specific internal sample control (ISC) RNA in serum, oral fluid, and fecal samples subjected to ≤15 freeze-thaw cycles.

Target	Specimen	Intercept*	Slope† (95% CI)	p
PRRSV	Serum	1.97	0.000 (−0.033, 0.033)	>0.05
	Oral fluid	1.00	−0.027 (−0.033, –0.020)	<0.05‡
	Feces	0.64	−0.005 (−0.017, 0.008)	>0.05
ISC	Serum	2.05	−0.003 (−0.019, 0.014)	>0.05
	Oral fluid	1.85	−0.052 (−0.085, –0.018)	<0.05‡
	Feces	1.50	−0.011 (−0.035, 0.012)	>0.05

PRRSV and ISC RT-rtPCR (Idexx) results were re-expressed as efficiency standardized Cqs (ECqs) and analyzed by a mixed-effects regression model.

^* Mean ECqs in untreated samples.

^† Mean ECq change per freeze-thaw cycle.

^‡ Significant effect of freeze-thaw at α = 0.05.

Figure 3.

Regression analyses and (mixed-effects regression model, MRM) showing the effect of freeze-thaw on (A) porcine reproductive and respiratory syndrome virus (PRRSV) RNA and (B) a porcine-specific internal sample control (ISC) in serum, oral fluid, and fecal samples exposed to ≤ 15 freeze-thaw cycles (study 2) and then tested with a PRRSV RT-rtPCR (Idexx). RT-rtPCR results were re-expressed as efficiency standardized Cqs (ECqs). Regression equations marked with an asterisk (*) indicate that freeze-thaw had a significant effect on PRRSV or ISC ECqs (MRM: p < 0.05).

In study 3, increasing PRRSV concentration had no effect on ISC ECqs in serum, oral fluid, or fecal samples (MRM: p > 0.05; Table 4). The highest PRRSV ECq values were obtained from the neat sample on each specimen (PRRSV MLV + 5-mL PRRSV-negative specimen), with 34.5, 35.9, and 44.7 ECqs for serum, oral fluid, and fecal samples, respectively. The type-III ANOVA showed no difference in effect between specimens (p > 0.05).

Table 4.

Increasing porcine reproductive and respiratory syndrome virus (PRRSV) concentration did not affect internal sample control (ISC) efficiency standardized Cq (ECq) values in serum, oral fluid, or fecal samples (study 3).

Increasing PRRSV concentration (ECqs)*	Change (Δ) in ISC ECqs
	Serum	Oral fluid	Feces†
0.0 (baseline)	0.00	0.00	0.00
0.5	0.00	0.00	0.00
1.5	0.00	0.00	−0.01
5.0	0.00	0.00	−0.03
10.0	−0.01	−0.01	−0.05
20.0	−0.01	−0.02	−0.11
35.0	−0.02	−0.03	−0.19
40.0	−0.03	−0.04	−0.22

^* Analysis of PRRSV and ISC RT-rtPCR (Idexx) efficiency standardized Cqs (ECqs) using a mixed-effects regression model found that PRRSV concentration had no effect on ISC ECqs (p > 0.05).

^† Fecal suspension created by suspending 10 g of feces in 50 mL of PBS (20% w/v suspension) and centrifuging at 3,300 × g for 3.5 h.

Discussion

We found that the effect of temperature × time and freeze-thaw cycles on the decay of detectable PRRSV and ISC RNA was specimen dependent, with serum being the most robust specimen type. Overall, our results on PRRSV temperature × time effect in serum agreed with reports describing RNA stability in serum specimens from patients infected with HCV (i.e., no significant difference in HCV quantification between immediate testing and testing after storage at 4°C for 5 d), but detected a decrease in HCV RNA levels in samples stored at room temperature for 5 d. ¹⁵ Other studies have also demonstrated the stability of HCV RNA in serum after storage at 25°C for up to 3 d, ⁹ and stability of human immunodeficiency virus in plasma samples maintained at 4°C and 21°C for up to 72 h. ³³ Cumulatively, these findings suggested that viral RNA is relatively stable in blood-derived specimens (e.g., serum, plasma), perhaps because serum proteins may have a protective effect on NAs.^4,13

On the other hand, we found that PRRSV and the porcine-specific ISC were less stable in oral fluids and fecal samples and declined continuously in ECqs at all temperature × time treatments. Although other data are limited, studies on porcine epidemic diarrhea virus and transmissible gastroenteritis virus (Alphacoronavirus 1) in fecal samples showed a decrease in detection of both targets at 4, 21, 36, and 45°C as storage time increased. ²⁰ In tissue, a decrease in DNA levels of African swine fever virus was observed over time in lungs stored at 4 and 23°C. ²⁴ Given the central role of laboratory testing in disease detection, control, and elimination, further studies are needed to understand the effect of storage temperature by time on specimens to be tested by rtPCR.

Retesting samples is a common practice in both diagnostic and research settings, either to confirm rtPCR results or to test for additional pathogens. However, this necessitates holding samples at storage temperatures over time or exposing samples to repeated freeze-thaw cycles. We detected both PRRSV and ISC RNAs in all serum, oral fluid, and fecal samples at all freeze-thaw treatments; however, we observed a decline in PRRSV and ISC ECqs in oral fluid samples. Studies assessing the effect of exposing clinical specimens to freeze-thaw cycles in viral RNA are mostly limited to human pathogens and blood-derived specimens in particular. In serum, no significant decrease in HCV level was found after 10 freeze-thaw cycles. ⁹ Similarly, dengue virus (Orthoflavivirus dengue) levels in serum were not affected after 5 freeze-thaw cycles. ¹ Although we found no information on the process(es) that affect viral NAs during the freeze-thaw process, our results indicate that serum and fecal samples are more protective of RNA than oral fluids.

Although serum samples used in our study came from pigs that were either vaccinated with a PRRSV MLV vaccine or inoculated with different isolates of wild-type PRRSV, the effect of each treatment did not vary (i.e., no isolate-specific differences were detected). Similarly, previous data on PRRSV temperature stability demonstrated no difference among viral isolates in the rate of inactivation of infectious virus (half-life) or in PRRSV RNA detection. ¹⁸ It should be noted that the effect of extrinsic factors in different specimen types described in our study are applicable for PRRSV, and additional studies will be required to test the effect of these conditions on other pathogens and other specimen types.

In routine rtPCR testing, standard procedures will vary somewhat among laboratories and discordant test results will occur as a result of variation among technicians, extraction and/or amplification protocols, reagents, equipment, etc.^14,34 Thus, quality controls (i.e., positive/negative extraction and amplification controls) are mandatory in routine RT-rtPCR testing to account for variation in the testing procedures and to ensure the accuracy of the results. In addition to these controls, ISCs have been used with increasing frequency in molecular research and routine testing to monitor sample quality. In basic research, the ISC is typically a reference gene that is inherent to the host-derived specimen and is consistently expressed regardless of the host’s sex, age, or disease status. We used a commercial RT-rtPCR for the simultaneous detection of PRRSV and a porcine-specific ISC to evaluate the effect of extrinsic factors on the detection of both NAs. We found that the concentration of PRRSV (1 × 10^-1–1 × 10^-6) did not affect the detection of the ISC in serum, oral fluid, or fecal samples. In our 3 studies, detection of the ISC was highly consistent among the specimens tested, which suggests that failure to detect the ISC indicates a problem at some point between sampling and testing. Under such circumstances, retesting and/or resampling is recommended.